Skip to Content
Fund Spy

Judging International Index Funds No Easy Task

Offerings can show big variations from one day to the next.

Index funds don't always match their benchmarks--or each other. That probably doesn't shock you. After all, in the past few years, Morningstar and other financial publications have drawn attention to  Vanguard Total Bond Market Index (VBMFX) and  Vanguard Short-Term Bond Index (VBISX)--each of which fell far short of its benchmark in 2002--and have noted the extent to which high fees on certain other index funds render it almost impossible for them to effectively track their chosen benchmarks.

But when bonds and high fees aren't involved and the funds are run by experienced index-tracking managers following a benchmark of large-cap stocks, it would seem their returns--both daily and over longer periods--would essentially mirror those of one other and the index. That's why I was surprised to see the opposite situation with the Fidelity and Vanguard funds attempting to track the MSCI EAFE Index of large caps in developed foreign markets.

Depends on What Day You Look
In early September, I checked the trailing 12-month returns of both funds to see how they had fared since Fidelity had cut the expense ratio on  Fidelity Spartan International Index  to 10 basis points (0.10%) from 47. That made its fund about 19 basis points cheaper than  Vanguard Developed Markets Index , rather than 18 basis points more expensive. The cut was put in place at the end of August 2004; a year's worth of data had just become available.

The search revealed that the Vanguard fund was ahead, 26.87% to 26.41%, for the trailing 12 months through Sept. 9, 2005. That 46-basis-point gap was noteworthy for index funds tracking the same index, given Vanguard's expense disadvantage. Yet that result wasn't shocking. Vanguard previously had demonstrated an ability, at both its domestic and international index-trackers, to make up more than a few basis points here and there by using its expertise in the nuances of the art.

Surprise did strike, though, when I looked again just two weeks later. The situation had reversed, with Fidelity suddenly showing a comfortable lead of more than 40 basis points in the 12-month trailing returns. Regardless of the imperfections of index-tracking, you wouldn't expect such a quick turnabout for two nearly identical portfolios. The shock intensified the next day. On Sept. 27, Fidelity's newly won lead suddenly disappeared, as the Fidelity fund lagged the Vanguard fund by a full 40 basis points, losing 0.70% to its rival's 0.30%. (EAFE was down 0.64%.) That one day wiped out Fidelity's entire lead, leaving the rival offerings with nearly identical returns for the 12-month period.

What's more, such wide daily divergences aren't unusual--and they can occur in either direction. On Oct. 10, for example, the Fidelity fund lost 0.21% while the Vanguard fund lost 0.51%. After that day, Fidelity again had a substantial lead in the 12-month trailing return. But take it with several grains of salt, for we now know that the comparison could look quite different by the time you read this.

What Gives?
There are several likely explanations for such wide variations in the two funds' daily--and yearly--returns. Two involve fair-value pricing--a topic we wrote about most recently in a July column.

First, funds have adopted different policies in deciding when to apply estimated prices rather than simply plugging in a stock's official price at the local market's close. That could well account for the wide divergence on Sept. 27. On that day, it seems that the Fidelity fund, whose return was close to EAFE's, did not use fair-value pricing, while Vanguard's offering did, or perhaps they both used it but Vanguard applied the process to more stocks than Fidelity did.

Second, even when both funds do decide to use fair-value prices, they can--and no doubt often do--assign different values to the same stocks. Many firms do use third-party services to value stocks, but even then, they can get different prices if they rely on different services. Fidelity uses an internal committee appointed by the Board of Trustees, while Vanguard uses a third-party service in combination with an internal fair-value committee that has the final say. And of course each firm sets its own policy as to when to implement fair-value pricing.

On the day of the London bombings, the Vanguard and Fidelity funds probably both used fair-value pricing, given that both reported daily returns that were significantly higher than EAFE's--yet the two funds were still 22 basis points away from each other. That was probably an instance where the funds ended up assigning different values to some stocks. (We must speculate to some extent, because while Fidelity and Vanguard spokesmen did provide helpful general information about their fair-value pricing policies in response to our queries, understandably they, like all fund companies, are reluctant to disclose specifics about their implementation of fair-value pricing.)

Finally, although probably of lesser importance in the size of the day-to-day variations, there's a third contributing factor involved here: It's tougher to track indexes than most people think, especially those indexes beyond the S&P 500. After all, the relative weights of each stock in EAFE change daily, and those many stocks--the two funds each own more than a thousand securities--trade in wide variety of markets in numerous time zones. Adjusting the portfolio each day to exactly match the performance of those stocks gets complicated. In fact, an EAFE tracker typically doesn't even try to own the precise correct amount of each stock. It will own the vast majority of them but then try to approximate the rest of the portfolio with the addition of futures, for example. No wonder different funds can have slightly different returns even when they're not using fair-value pricing.

A Sticky Situation That Demands Attention
As we noted in the July column as well as in one written back in 2001, while we generally support the idea of fair-value pricing as a weapon to combat market-timing, we think its usage raises important concerns. When funds are adjusting their own NAVs and rival funds are assigning different prices to the same stocks, strict oversight by regulators is required. Note that in most cases mentioned in this column--unlike the instance of the London bombings--the markets were not undergoing abnormal stresses. This indicates that this issue is a daily concern among international funds, not something that only pops up on rare occasions.

Some may say I am overstating the case, for investors shouldn't be picking funds on the basis of one-year returns anyway. That may be true for actively managed funds. But with index trackers, one should be able to rely on a year's worth of data to provide an indication of how they stack up against competitors, how effectively they're tracking their benchmarks, and whether expense differences make one fund preferable to another. When trailing-returns figures are rendered unreliable for reasons that have nothing to do with the funds' index-tracking skills or expenses, then we've got a problem.

Unfortunately, there's no simple answer to this dilemma. We provided some ways to deal with the issue in the previous columns mentioned above, and we'll do so again in the future. For now, what's clear is that this is a topic that should not be ignored.

Sponsor Center